Practical Bioinformatic DNA-Sequencing Pipeline for Detecting Oncogene Amplification and EGFRvIII Mutational Status in Clinical Glioblastoma Samples.

ABSTRACT

Glioblastoma is a malignant brain tumor with dismal prognosis. Oncogenic mutations in glioblastoma frequently affect receptor tyrosine kinase pathway components that are challenging to quantify because of heterogeneous expression. EGFRvIII, a common oncogenic receptor tyrosine kinase mutant protein in glioblastoma, potentiates tumor malignancy and is an emerging tumor-specific immunotarget, underlining the need for its more accessible and quantitative detection. We used normalized next-generation sequencing data from 117 brain and 371 reference clinical tumor samples to detect focal gene amplifications across the commercial Ion AmpliSeq Cancer Hotspot Panel version 2 and infer EGFRvIII status based on relative coverage dropout of the gene's truncated region within EGFR. In glioblastomas (n = 45), amplification of EGFR [18 (40%)], PDGFRA [3 (7%)], KIT [2 (4%)], MET [1 (2%)], and AKT1 [1 (2%)] was detected. With respect to EGFR and PDGFRA amplification, there was near-complete agreement between next-generation sequencing and in situ hybridization. Consistent with previous reports, this method detected EGFRvIII exclusively in EGFR-amplified glioblastomas [8 (44%)], which was confirmed using long-range PCR. Our study offers a practical method for detecting oncogene amplifications and large intragenic mutations in a clinically implemented hotspot panel that can be quantified using z scores. The validated detection of EGFRvIII using DNA sequencing eliminates problems with transcript degradation, and the provided script facilitates efficient incorporation into a laboratory's bioinformatic pipeline.